Reference no: EM134008593
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
Assignment - Analysis of Exam Timetabling Problem
Objectives
Reinforce knowledge and understanding of heuristic and metaheuristic search techniques covered in Weeks 5-6. The concepts discussed in Artificial Intelligence and Algorithms For Problem Solving are highly relevant to this assessment.
Analyse a complex real-world problem and evaluate the suitability of different search approaches.
Assessment relates to Unit Learning Outcomes 1, 2 and 4.
Problem Description
Universities must generate examination timetables that:
Avoid clashes for students enrolled in multiple units
Consider room availability
Consider staff scheduling constraints
Ensure fairness in exam distribution
Complexities include:
Combinatorial explosion of possible timetables
Conflicting constraints
No single optimal solution satisfying all requirements
Increasing numbers of units, students, and constraints make exhaustive or optimal search impractical
Case Scenario:
As a key member of the Timetabling Team, analyse this problem and propose suitable search approaches. Concepts from Data Structures Algorithms may support the analysis of efficient search strategies and scheduling methods.
Assignment Tasks
Section 1: Problem Characterisation
Describe the exam timetabling problem in your own words
Explain what is being optimised or solved
Identify key constraints and sources of complexity
Justify why the search space is large or difficult to explore
Section 2: Limitations of Classical Search
Explain why classical search methods such as Breadth-First Search (BFS) or Depth-First Search (DFS) are not suitable
Discuss limitations in terms of:
Time complexity
Space requirements
Scalability
Practical feasibility
Link explanations clearly to the exam timetabling scenario
Section 3: Heuristic or Metaheuristic Approach
Apply one or more suitable methods discussed in Week 5 and Week 6
Justify why the chosen method(s) are appropriate
Explain:
Representation of candidate solutions (timetables)
How the search is guided toward better solutions
How constraints are handled
Knowledge from Computer Science and optimisation-related algorithm design can assist in evaluating heuristic and metaheuristic techniques for solving scheduling problems.
Section 4: Trade-offs, Risks, and Limitations
Discuss trade-offs involved in the proposed approach
Consider:
Solution quality vs computational cost
Lack of optimality guarantees
Risks such as local optima or slow convergence
Explain situations where the approach may perform poorly
Section 5: Reflection on Practical Use
Discuss how the proposed approach may be used in practice
Consider whether it would be used alone or combined with other methods
Identify additional information or constraints affecting implementation
Use of Literature
Use relevant publications from open literature where appropriate
References must directly support arguments
All sources must be clearly and correctly cited
Excessive or irrelevant referencing will not improve marks
Proper citations and reference list
Word limit: Maximum 1,000 words (including references)
APA 7 referencing required